Function: Basic N1QL Prepared Select Statment


    Goal: Iterate through a basic N1QL SELECT where Eventing interacts with the Data service via a prepared N1QL statement.

    • This function basicN1qlPreparedSelectStmt demonstrates how use a prepared N1QL SELECT statement with a passed parameter.

    • Typically, this is done for greater performance, the cluster will typically not prepare a statement if it is already prepared.

    • We have just one positional parameter $1 for "doc.iata", if we had a second parameter we would use the placeholder $2, and so on. Note positional parameters are passed in an array.

    • A commented out example of using named parameters is also shown. Note named parameters are passed and an object.

    • Requires the "travel-sample" sample dataset to be loaded.

    • Requires Eventing Storage (or metadata collection) and a "source" collection of travel-sample.inventory.airline.

    • Deploy the Function with a Feed Boundary "From now" (Note you will log 187 lines if you use "Everything").

    • Assuming you deployed "From now" mutate any document in "travel-sample" to generate a log line.

    • For additional details refer to N1QL Statements and The N1QL() function call

    • [Optional] if Feed Boundary is "Everything" you can use SQL++ (N1QL) to add an index for performance:

      • CREATE INDEX adv_airline_type ON default:`travel-sample`.inventory.route(airline) WHERE (type = 'route')

    • basicN1qlPreparedSelectStmt

    • Input Data/Mutation (via the following N1QL statement)

    • Output Data/Logged

    // To run configure the settings for this Function, basicN1qlPreparedSelectStmt, as follows:
    // Version 7.0+
    //   "Listen to Location"
    //     travel-sample.inventory.airline
    //   "Eventing Storage"
    //     rr100.eventing.metadata
    //   Binding(s) - none
    // Version 6.X
    //   "Source Bucket"
    //     travel-sample
    //   "MetaData Bucket"
    //     metadata
    //   Binding(s) - none
    function OnUpdate(doc, meta) {
        // ignore information we don't care about
        if (doc.type !== 'airline') return;
        var route_cnt = 0;       // we want to get the total routes per iata
        // positional parameter(s)
        var results = N1QL("SELECT COUNT(*) AS cnt " +
            "FROM `travel-sample`.`inventory`.`route` " +
            "WHERE type = \"route\" AND airline = $1",
            [doc.iata], { isPrepared: true }
        // named parameter(s)
        var max_dist = 120;
        var results = N1QL("SELECT COUNT(*) AS cnt " +
            "FROM `travel-sample`.`inventory`.`route` WHERE type = $mytype " +
            "AND airline = $myairline AND distance <= $mydistance",
            { '$mytype': 'route', '$mydistance': max_dist, '$myairline': doc.iata },
            { 'consistency': 'none', isPrepared: true }
        for (var item of results) {   // Stream results using 'for' iterator.
            route_cnt = item.cnt;
        results.close();              // End the query and free resources held
        // Just log the KEY, AIRLINE and ROUTE_CNT it to the Application log
        log("key: " + + ", airline: "+doc.iata+", route_cnt: "+route_cnt);
    UPDATE `travel-sample`.`inventory`.`route` USE KEYS "airline_24" SET id = 24;
    2021-07-19T07:48:10.810-07:00 [INFO] "key: airline_24, airline: AA, route_cnt: 2354"